Answer
Long term predictions are extrapolations. Why they are not good is explained in exercise 50.
The data set may have "holes" in it, and we may wish to estimate a value within the range of the dataset, that has not been given in the dataset, This is called interpolation.
Also, the model may be useful to estimate the next period, immediately after the last in the data set. In this way the extremes mentioned in exercise 50 are not yet encountered, and we may take these estimates as being realistic.
Work Step by Step
No steps involved. The answer contains the needed explanations.
